Charge injection based electrical stimulation on polypyrrole planar electrodes to regulate cellular osteogenic differentiation

In this study, polypyrrole (Ppy) electrodes were prepared to support an electrical stimulation to MC3T3-E1 cells for regulating their osteogenic differentiation. The charge injection capacity (CQ) of the Ppy electrodes could be adjusted by the Ppy thickness, and a higher CQ could make the electrode able to produce a higher charge injection quantity (Qinj) at applied voltage. The Qinj onto electrode could be considered as the intensity of the stimulation pulse to cells, and the pulse frequency means the number of electric stimulation with Qinj at one second. Hence, we conducted the present work in the view of Qinj. When the cells were electrically stimulated for 1 hour per day, the electrodes with Qinj ranged in 0.08–0.15 μQ had an obvious role in enhancing cellular osteogenic differentiation whereas Qinj of lower than 0.03 μQ or more than 0.30 μQ gave the stimulations with no or negative effects. And the stimulation with 1 or 25 Hz showed to enhance the differentiation, whereas the stimulation with 50 Hz gave an inhibiting effect. We further found the osteogenic differentiation potential triggered by electrical simulation was related to cell growth stage, and the stimulation carried out at early stage (day 2–5) during 8 days cell culture showed more contribution to enhancing osteogenic differentiation than that at later stage (day 6–8). It is proposed that the desired stimulation effects require that an appropriate voltage-gated calcium ion channel and efficient intracellular calcium ion oscillation are well activated. This work therefore reveals Qinj as an important electrode parameter to decide effective simulations and provides an insight into understanding of the role of electrode material characters in regulating cellular osteogenic differentiation during stimulation.


Introduction
Electric elds are believed to play a critical role in major biological processes such as embryogenesis, muscle contraction, wound healing and tissue regeneration. [1][2][3] Inspired by the natural electrical properties of bone, an electric eld stimulation generated by exogenous electric devices demonstrates a great potential to regulate osteogenic functions of stem and osteoblast-like cells and promote bone growth. [4][5][6][7][8][9][10] To generate an effective electrical stimulation, the electric devices/ electrodes and material selections are crucial. Compared with traditional parallel electrodes, planar interdigitated electrodes (IDE) integrate the pairing electrodes onto the same plane and generate stimulation under low voltage and in a direct, effective, highly reproducible and controlled manner. [11][12][13] Various materials have been served as electrodes with different electric eld intensities (or voltages) for electrical stimulation in osteogenic functions. For instance, McCullen et al. used a gold electrode with interdigitated shape to deliver stimulation under alternating electric eld of 1 V cm À1 , and the results showed an increase in cellular osteogenic differentiation potential of human adipose-derived stem cells (hASCs). 4 An electrode material of conductive 3D silk foams produced stimulation with an interpenetrating network of poly (pyrrole-co-(2hydroxy-5-sulfonic aniline)) bone tissue scaffolds with the electric eld of 1 V cm À1 , and had a potential to enhance the differentiation toward osteogenic outcomes of hMSCs. 14 A Ti electrode covered by dened nanotubular TiO 2 produced stimulation with constant electric eld strength of 200 or 400 mV cm À1 , which triggered the cellular osteogenic induction of mesenchymal stem cells (MSCs). 5 An electrode material of conductive polypyrrole lms supported stimulation with electric eld of 350 mV cm À1 , and showed its capability to promote osteogenesis of calcium deposition in the extracellular matrix of rat bone marrow mesenchymal stem cells (rBMSCs). 15 Pelto et al. showed an enhanced early osteogenic differentiation of adipose stem cell proliferation on an electrode material of chondroitin sulfate (CS)-doped polypyrrole-coated polylactide scaffolds, which provided stimulation with DC voltage of 200 mV cm À1 . 16 Hardy et al. developed electrode materials of biomineralized conducting polymers and conductive polycaprolactone (PCL) ber-based bone tissue scaffolds to enhance the differentiation towards osteogenic outcomes of hMSCs with the electric eld of 100 mV cm À1 . 17,18 A moderately conductive articial extracellular matrix coated polyaniline substrates was employed by Thrivikraman to apply stimulation with intermittently electric eld of 3.6 mV cm À1 , and the data showed an enhancement in osteogenic differentiation potential of human mesenchymal stem cells (hMSCs). 6 It is evident from this vast literatures that a large range (1 V to 3.6 mV cm À1 ) of the applied voltage intensities were involved to stimulate to cells for osteogenic functions due to different electrode materials.
An electrode for electric simulation to cells is desired to elicit effective cellular responses at applied voltages as low as possible in order to avoid electrolysis and accompanying pH-and ionicgradients in long-term experiments. 5,19,20 For these electrodes, electrical properties would be of greater concern. 21-23 The interaction ability of the electrodes with external species is believed to be proportional to electrical property of charge injection capacity (C Q ) of electrodes at a given applied voltage. 24 Therefore, it is pursued for a material to support an electrode with high C Q .
In this study, conductive polymer polypyrrole (Ppy) was selected as electrode material due to its reversible-faradaic reactivity which resulted in a high C Q . Ppy planar interdigitated electrodes were prepared by electro-polymerization, and the C Q of electrodes was adjusted by changing Ppy polymerized conditions. An electrical stimulation with biphasic pulse mode was performed during culturing of MC3T3-E1. In the view of charge injection, the effect of electrical stimulation on the cellular osteogenic differentiation was studied at different applied voltages and stimulation times (the pulse frequency and the total stimulation time). Moreover, the possible mechanism of electrical stimulation to cells was also discussed.

Preparation of Ppy electrodes
ITO glass (10 mm Â 10 mm) was patterned by laser etching to form 500 mm width ITO strip and spacing (Fig. 1A). The electropolymerization of Ppy was conducted with a two-electrode electrochemical system under galvanostatic conditions. The patterned ITO substrate and platinum sheet were used as a working and counter electrodes, respectively. The current densities of 3 mA cm À2 , 5 mA cm À2 and 8 mA cm À2 with 30 s were utilized to electro-polymerize for forming Ppy electrodes with different Ppy thickness. As-formed electrodes were washed with deionized water and followed by air-drying overnight. The electro-polymerization solution was consisted of 0.1 M pyrrole (Sigma-Aldrich, USA) and 0.1 M sodium p-toluene sulfonic acid (pTS) (Macklin, China).

Characterizations of Ppy electrodes
Raman spectra of the prepared electrodes on patterned ITO substrate were taken by OMNIC Dispersive Raman (Thermo Fisher Scientic, USA) with a DXR laser operating at 532 nm with incident power of 10 mW. The prepared electrode materials were peeled from patterned ITO substrate and mixed with KBr, and then FTIR spectrum was collected in the range of 4000-1000 cm À1 . The thickness and morphologies of the prepared electrodes were examined by Field Emission Scanning Electron Microscopy (FESEM; Hitachi, SU-70). Prior to SEM analysis, all prepared electrodes were sputter-coated with a thin layer of gold.

Electrochemical analysis
All electrochemical experiments were conducted by two-electrode system in culture medium using a computer-controlled CHI 660D electrochemical workstation (Chenhua Instrument Co., Shanghai, China). The culture medium was consisted of alpha-modied Minimum Essential Medium (Alpha-MEM, Gibco, Waltham, MA) and 10% fetal bovine serum (Gibco) and 1% antibiotics (10 000 units per mL penicillin and 10 000 mg mL À1 streptomycin, Gibco). As a planar interdigitated electrode, one side of Ppy electrode was used as working electrode and another as counter electrode. Cyclic voltammetric (CV) measurements were performed from 1 V to À1 V at a scan rate of 50 mV s À1 . The current-time curves were collected under a biphasic pulses signal (used in electrical stimulation experiments) with different voltages. Then the current was integrated over time in the charged period in each curve using Originpro 8.5 soware. The obtained integral area was the Q inj (the total amount of charge injected on the electrode during a stimulus pulse) on electrode (Fig. S2 †). 25 Finally, Q injvoltage curves were drawn according to the calculated values.

Cell spreading, viability and proliferation
Ppy electrode was placed in a home-made culturing device (made by polytetrauoroethylene, shown in Fig. S1 †), and then Mouse calvaria-derived pre-osteoblastic cells (MC3T3-E1) were seeded at a density of 2 Â 10 4 cells per cm 2 and maintained at 37 C and 5% CO 2 in a humidied atmosphere.
Cellular attachment and spreading on Ppy electrodes aer 1 day of culture was determined by using immunouorescence staining. In addition, typical morphology of cells on Ppy electrodes aer 2 days of stimulation was also observed. The cells on the electrodes were xed with paraformaldehyde (4% in PBS) for 15 min, and then permeabilized with 0.4% Triton X 100 for 15 min, blocked in solution contained BSA and FBS (2% in PBS) for 1 hour. Aer washing in PBS, cells were incubated with uorescent dye of anti-vinculin (EPR8185, Abcam, UK), rhodamine phalloidin (Phalloidin-iFluor™ 594 Conjugate, AAT Bioquest, Inc. USA) and 4 0 ,6-diamidino-2-phenylindole (DAPI, ENZ-52404, Enzo Life Sciences, Switzerland). The vinculin (green), cytoskeleton (red) and nucleus (blue) were visualized by confocal laser scanning microscopy (Zeiss LSM 780, Germany).
The viability and proliferation of the cells cultured on Ppy electrodes were determined by the Cell Counting Kit-8 (CCK-8, Dojindo Laboratories, Kumamoto, Japan). Briey, cells on Ppy electrodes were transferred to a new 24-well plate aer culture for 1 and 5 days, and then 500 mL fresh culture media and 50 mL of CCK-8 solution were added to each well. Aer incubated for 3 h at 37 C, the solution was dispensed into a 96-well plate, and colorimetric measurements of formazan dye were conducted with the microplate reader at 450 nm.

Electrical stimulation
Before seeding cells, Ppy electrode was placed into a homemade culturing device (Fig. S1 †), and then two platinum wires were xed onto the two conductive sides in an electrode by polytetrauoroethylene screws. MC3T3-E1 cells were seeded at a density of 2 Â 10 4 cells per cm 2 , and all stimulations to cells were performed aer 1 day pre-culture to permit cellular attachment and spreading. The two wires were connected to a waveform generator (DG1022 type, Rigol Electronic Co., Ltd., Beijing, China) and the stimulation was applied on cells under a biphasic pulse with 10 ms pulse duration.
The Ppy electrodes were conducted to stimulate cells under the applied voltages of 5 mV or 25 mV at 1 Hz, trying to understand the role of Q inj . The electrodes with the same Q inj were conducted to stimulate cells under different voltages calculated by charge-voltage curves. The stimulation frequencies of 1 Hz, 25 Hz or 50 Hz with the applied voltages of 5 mV were adopted to stimulate cells on the Ppy electrodes. The cell culture period of 8 days was divided into two stimulation stages, the early stage (day 2-5) and later stage (day 6-8). The stimulation to cells on the electrodes at the two stages was performed with applied voltage of 5 mV. The cells on the Ppy electrodes without electrical stimulation (ES) were used as control (non-ES).

Alkaline phosphatase analysis
Aer 8 days of culture, culture medium was removed, and the cells on Ppy electrodes were rinsed with PBS for three times. The cells were lysed with CelLytic Buffer (Sigma, St. Louis), and the obtained cell lysate was centrifuged with the speed of 12 000 rmp at 4 C for 15 min. The supernatants were assayed by LabAssay™ ALP (Wako Pure Chemical Industries, Ltd. Japan) via measuring the optical density at a wavelength of 405 nm. The ALP activities were obtained by normalizing the quantitative assay values to total protein contents tested in a BCA protein assay.

Measurement of intracellular Ca 2+
Intracellular Ca 2+ levels were measured using Fluo4-AM calcium indicator (Dojindo Laboratories, Japan) aer 2 days of stimulation or stimulation at later stage (day [4][5]. Aer stimulation, the cells on Ppy electrodes were rinsed with HBSS and the 400 mL of Fluo4-AM dye loading solution (4 mM in HBSS) was added. The plates were incubated at 37 C for 30 minutes for the complete reaction. Successively, the cells were washed with HBSS and collected by 0.5 mL trypsin. Aer three times of centrifugation, mean uorescence intensity of cell suspension was measured by ow cytometer (Cytoex, Beckman Coulter, China) with the emission wavelength at 518 nm and exaction wavelength at 495 nm.

Statistical analysis
All the quantitative data in this study were expressed as mean AE standard deviation (S.D.). Statistical analysis was performed using the soware of Statistical Package for the Social Sciences (SPSS) version 19. The signicant differences were performed by one-way analysis of variance (ANOVA) followed by Tukey's test. In all of the statistical evaluations, the differences with p < 0.05 were considered as statistically signicant.

Characterizations of Ppy electrodes
As shown in Fig. 1A, the patterned electrode was interdigitated shape (10 mm Â 10 mm) with 500 mm width and spacing. Fig. 1B shows the photograph of patterned ITO substrate (le) and the prepared electrode (right) that obtained by electrochemical polymerization. Obviously, black materials with interdigitated shape were found aer polymerization.
As shown in Fig. 1C, the obtained electrode exhibits typical Raman peaks at 1580 and 980 cm À1 , which are assigned to the symmetric C]C backbone stretching mode and C-H plane deformation. [26][27][28][29] The FTIR spectrum (Fig. 1D) shows the broad band around 3000-3500 cm À1 , which could be ascribed to N-H and aromatic C-H stretching vibrations. The two absorption peaks at 2851 cm À1 and 2922 cm À1 are corresponded to symmetric and asymmetric stretching vibrations of C-H groups, respectively. The absorption peaks at 2360 cm À1 is related to stretching vibrations of C-N. The bands at 1635 cm À1 and 1540 cm À1 are corresponded to C]C ring stretching vibration of pyrrole ring and benzene ring of dopant (pTS). [30][31][32] The Raman and FTIR spectra of the prepared electrode indicated that Ppy was formed on the patterned ITO substrate.
The SEM images of the electrode cross-section (Fig. 1F-H) show that the thickness of the Ppy electrodes on patterned ITO substrate (Fig. 1E) by electro-polymerization increased with current densities. The electrode thickness of 20 AE 2 nm, 50 AE 3 nm and 80 AE 5 nm corresponded to the electro-polymerization at current densities of 3 mA cm À2 , 5 mA cm À2 and 8 mA cm À2 , respectively. Fig. 1J-L shows the morphologies of the obtained electrodes at different current densities. Ppy grains were observed to homogeneously cover the patterned ITO substrate (Fig. 1I) at current density of 3 mA cm À2 (Fig. 1J), and the grains became large at current density of 5 mA cm À2 (Fig. 1K). Further increasing the current density to 8 mA cm À2 , the large grains demonstrated a smoother and denser surface morphology (Fig. 1L). The Ppy electrodes prepared at current density of 3 mA cm À2 , 5 mA cm À2 and 8 mA cm À2 were named by Ppy-I, Ppy-II and Ppy-III, respectively.

The C Q and Q inj of Ppy electrodes
The CV curves ( Fig. 2A) demonstrate that three typical electrodes exhibit different loops within the voltage window ranging from 1 to À1 V in culture medium. The enclosed area in CV curve with the overall voltage range represents charge injection capacity (C Q ) of the electrodes, 33 and C Q increased with polymerized current densities. Obviously, the patterned ITO substrate demonstrated a much lower C Q compared to that of Ppy electrodes. Fig. 2B shows the electric current responses of Ppy-I, Ppy-II and Ppy-III electrodes to the electric voltage pulses at 5 mV. The curve area of time integral of the current in charging period was considered to be the quantity of electric charge injected This journal is © The Royal Society of Chemistry 2018 onto the electrodes, i.e. Q inj (Fig. S2 †). 25 Fig. 2C shows the Q inj obviously increased upon the increasing of electrode C Q and applied voltage.

Cell adhesion, spreading, viability and proliferation on Ppy electrodes
Fig. 3A-C shows the cellular morphologies on the prepared electrodes indicated by immunouorescent staining aer 1 day of culture. F-actin and vinculin indicated the cytoskeletonrelated proteins associated actin and integrins, respectively. The organization of actin and expression of vinculin throughout the spreading regions of cells grown on all Ppy electrodes conrmed the cytocompatibility of the electrodes. The results revealed that the prepared Ppy electrodes were suitable for cell adhesion and spreading. Fig. 3D shows that cells on different electrodes presented the similar viability aer 1 day of culture and cellular proliferation increased signicantly aer 5 days of culture. These results indicated the excellent adhesion and biocompatibility of the electrodes.

The ES effects of stimulation intensity on osteogenic differentiation
For electrical stimulation to cells, 5 mV and 25 mV for applied voltages were selected because the voltages could give the Ppy electrodes to have signicant difference of Q inj in three orders of magnitude according to the Q inj -voltage curve (Fig. 2C).
As shown in Fig. 4A, at the applied voltage of 5 mV, Ppy-I electrode with Q inj of 0.01 mQ showed no effect on osteogenic differentiation due to low C Q . Ppy-II electrode with Q inj of 0.08 mQ expressed a promoted differentiation, while Ppy-III electrode with Q inj of 0.30 mQ inhibited the differentiation due to high C Q . At the applied voltage of 25 mV, Ppy-I electrode with Q inj of 0.03 mQ still showed no effect on osteogenic differentiation, whereas Ppy-II electrode with Q inj of 0.40 mQ and Ppy-III electrode with Q inj of 1.5 mQ inhibited differentiation. Those indicate the Ppy electrode with high C Q at low applied voltage is able to enhance differentiation. Fig. 4B-D shows the osteogenesis-related gene expressions of the cells stimulated on Ppy-II electrode with Q inj of 0.08 mQ (5 mV) and 0.30 mQ (25 mV). The results showed that the gene expressions of ALP and Runx2 were up-regulated for Q inj of 0.08 mQ, whereas the simulation with Q inj of 0.30 mQ showed no effect on ALP and Runx2 expression, and down-regulated the expression of OCN signicantly.
The effect of Q inj on differentiation on Ppy-II electrode was shown in Fig. 5A. ALP activity increased gradually with Q inj increasing, and Q inj of 0.08 mQ showed the largest ALP activity. The Q inj of 0.4 mQ decreased the ALP activity signicantly.
When Q inj of 0.08 mQ was set for other two Ppy electrodes using different applied voltages, Ppy-I (60 mV) and Ppy-III (1 mV) electrodes presented the same stimulation effect with Ppy-II electrode (5 mV) for enhancing osteogenic differentiation (Fig. 5B). Obviously, Ppy-II electrode with Q inj of 0.08 mQ (5 mV) showed the most signicant upregulation of osteogenic differentiation compared with Ppy-I and Ppy-III electrodes (Fig. 5A), which was selected for further studies.

The ES effects on cell cytoskeleton
When cells were seeded onto the Ppy electrode, the cells grew to homogenously distribute on whole electrode including conducting Ppy and spacing glass strips aer 2 days of stimulation, this could result from that all cells on the electrode received the equal stimulation effect. Fig. 6 shows the cellular cytoskeleton on the Ppy-II indicated by immunouorescent staining aer 2 days of stimulation under Q inj of 0.08 mQ (5 mV). The spreading area of actin and expression of vinculin in stimulation group was larger than that of nonstimulated group. The results indicated that electrical stimulation promoted the cellular spreading and migration on Ppy electrode.

The ES effect of stimulation time on cellular osteogenic differentiation
For electrical stimulation to cells at a biphasic electric pulse mode, the frequency of electric pulses means a stimulating pulse number in one second, and total stimulation time represents a stimulating day number in the whole cell culture period.
The effect of stimulating frequency on osteogenic differentiation for 1 hour stimulation per day was showed in Fig. 7A, 1 Hz or 25 Hz demonstrated to enhance the differentiation, whereas 50 Hz inhibited the differentiation. Fig. 7B-D shows the osteogenesis-related gene expressions under the stimulation with different frequency, the cellular gene expressions of ALP, Runx2 and OCN stimulated with 1 Hz or 25 Hz were upregulated, especially OCN expression. However, the stimulation with 50 Hz showed no effect on ALP and Runx2 expressions, but down-regulated OCN expression.
The electrical stimulation carried out at early stage (day 2-5) showed more signicant contribution to promotion of cellular osteogenic differentiation than that at early stage (day 6-8) in 8 day culture period (Fig. 8A), and a similar ALP level to that for 8 days.
The osteogenesis-related gene expressions of ALP (Fig. 8B), Runx2 (Fig. 8C) and OCN (Fig. 8D) that stimulated at early stage (day 2-5) were up-regulated. The stimulation carried out at later stage (day [6][7][8] showed no up-regulation of ALP gene expression, up-regulation of Runx2 and down-regulation of OCN. For the stimulated cells, their CaM gene expression showed no signicant change, and CaN gene expression was upregulated. As the last downstream gene of Ca 2+ signal pathway, NFAT expression of the cells that stimulated with Q inj of 0.08 mQ and 25 Hz or at early stages showed a signicant upregulation. Fig. 10 shows the intracellular Ca 2+ concentration indicated by the mean uorescence intensity (MFI) aer 2 of stimulation or stimulation at later stage (day 4-5) on cell on Ppy-II with Q inj of 0.08 mQ at the applied voltage of 5 mV. Fig. 6 Cellular morphologies on the Ppy-II were indicated by immunofluorescent staining after 2 days of stimulation (0.08 mQ, 5 mV). Cells were stained for the FA protein vinculin (green), cellular nuclei (blue) and actin cytoskeleton (red). The images share the same scale bar of 50 mm.  The ow cytometry assay showed that intracellular Ca 2+ concentration increased with culture time from 2 days to 5 days without stimulation. And aer 2 days of stimulation, intracellular Ca 2+ concentration increased signicantly, whereas stimulation at later stage did not affect the Ca 2+ concentration.

Discussion
The Ppy electrodes showed that their thickness increased with electro-polymerization current densities (Fig. 1). As a result, higher thickness led the electrode have higher C Q ( Fig. 1 and 2A) due to more spaces to accommodate electric charges derived from irreversible faradaic reaction. 34 For a given applied voltage, the electrode with higher C Q had a higher response to an applied voltage pulse (Fig. 2B) as well as a higher charge injection quantity (Q inj ) (Fig. 2C) than those with lower C Q . Q inj of the electrodes could be also tuned by the applied voltage. The electrodes with higher Q inj means to induce a stronger electrical simulation pulse intensity to cells (Fig. 2C), and could give a stronger stimulation effect at lower applied voltage.
During electrical stimulations to cells, electric elds, currents and electrochemical factors derived from electrodes will contribute to the stimulation. In this study, we focused on the effects of electric eld because there were no electrical current ow in Ppy stripes of the electrode under the present applied voltage and no redox peaks observed (Fig. 2A).
The Ppy lm was testied to be very stable on the patterned ITO substrate for soaking in culture medium and showed to well support cellular adhesion, spreading and proliferation, good cytocompatibility (Fig. 3). The assay of cellular osteogenic differentiation (Fig. 4A) showed that the Ppy electrodes with higher C Q produced signicantly upregulated effect at lower applied voltage as expected.
Since the stimulation pulse intensity is directly related to Q inj , the stimulation effects of the electrodes could be evaluated by their Q inj . The result (Fig. 4A) showed that no or negative simulation effect on cellular osteogenic differentiation was generated for the Ppy electrode with Q inj of low than 0.03 mQ (Ppy-I at 5 mV) or more than 0.30 mQ (Ppy-III at 5 mV), while the Ppy electrode with Q inj of 0.08 mQ (Ppy-II at 5 mV) produced the simulation to have an obvious role in enhancing osteogenic differentiation as well as the up-regulated the expression of osteogenesis-related genes (Fig. 4B-D). When the Ppy-I or Ppy-III electrode was applied by 60 mV or 1 mV voltage to have Q inj of 0.08 mQ, the two electrodes also demonstrated the obvious stimulation effect to enhance the differentiation (Fig. 5B). Hence, the Ppy electrode with higher C Q and the stimulation pulse intensity of Q inj of 0.08 mQ were appropriate to generate effective stimulations to cells for attaining the enhancement of osteogenic differentiation. In addition, this electrical stimulation promoted the cellular spreading and migration on electrode (Fig. 6).
For the electrical stimulation to cells, the stimulation time is another important parameter besides stimulation intensity of Q inj . The time could be divided into three parts, the time at one second with pulse number (the stimulation pulse frequency), the time at one day with stimulating hour (the daily stimulation time), and the time at the whole cell culture period with stimulating days (total stimulation time). The role of the daily stimulation time had been discussed in our previous work, 11 and 1 hour per day showed appropriate in promoting cellular osteogenic differentiation, and was adopted in this study.
When Q inj of Ppy-II electrode was set at 0.08 mQ (5 mV), the stimulation pulse frequency and total stimulation time were changed, the simulation pulse with 1 Hz or 25 Hz showed to enhance the differentiation, whereas that with 50 Hz inhibited the differentiation and osteogenesis-related gene expressions (Fig. 7). During cell culture for 8 days, the electrical stimulation carried out at early stage (day 2-5) demonstrated more obvious role in promoting cellular osteogenic differentiation than that  at later stage (day 6-8) (Fig. 8). This suggests that the early stimulation could make more signicant contribution in enhancing the differentiation in 8 day cell growth.
The assay of CaN and NFAT (Fig. 9) showed that their changing tendency was almost same with one of the electrical stimulation dependent osteogenic differentiation activity (Fig. 4, 7 and 8), revealing that the electrical simulation derived up-regulation in osteogenic differentiation could be attributed to that the calcium ion signaling pathway was activated to trigger osteogenic differentiation.
Based on the mechanism of electrical stimulation on cellular function, the electric eld of stimulation manipulates transmembrane potentials of cells and activates the voltage-gated calcium ion channels (VGICs). As a result, the uxes of calcium ions through the activated VGICs induce the intracellular calcium ion oscillation, and the osteogenic differentiation is triggered. [35][36][37][38] The VGICs have been proved to play an important role in enhancement of AD-MSCs' and osteoblasts' functions of osteogenic differentiation during electrical stimulation. 36 VGICs include high voltage-activated (HVA) channels that sub-classied into L-, N-, P-, Q-and R-types and low voltageactivated (LVA) channels (or T-type). VGICs to produce the biological function usually require specic stimulation voltage correspond to specic cells LVA channels are activated by lower membrane potentials, and HVA channels are opened in response to large membrane depolarizations and are activated at higher membrane potentials. 39,40 In this study, the electrical simulation generated by Q inj of 0.08 mQ showed the optimum effects in osteogenic differentiation, which might be attributed to that the studied Q inj range from 0.03 mQ to 0.30 mQ for Ppy electrodes involves activation of different VGICs, and the stimulation with Q inj of 0.08 mQ just activates a VGIC which can trigger well the osteogenic differentiation. The simulation to cells with high frequency of 50 Hz showed no role in promoting osteogenic differentiation (Fig. 7), it is speculated that the VGICs is activated too frequently to induce effective intracellular calcium ion oscillation, resulting in no activation of subsequent CaN/NFAT signaling pathway (Fig. 9). Since cell growth experiences adhesion, skeletal spreading, proliferation and differentiation, etc., we suggest that the cells at different growth stage could have specic stimulation sensitivity to VGIC activation and intracellular calcium ion oscillation. The obvious enhancement in intracellular Ca 2+ concentration at early stage may activate calcium ion pathway, whereas no effect on Ca 2+ concentration with stimulation at later stage (Fig. 10). The cells at early growth stage could have good sensitivity and result in the early stimulation with more contributions to promoting cellular osteogenic differentiation.

Conclusion
Charge injection capacity (C Q ) of polypyrrole (Ppy) electrodes could be tuned by electro-polymerization controlled Ppy thickness. The charge injection quantity (Q inj ) onto the electrodes was a vital parameter and considered as stimulation pulse intensity to stimulate cells. The electrodes with Q inj of 0.08 mQ resulted in the stimulation with a good effect in enhancing cellular osteogenic differentiation. The stimulation with 1 or 25 Hz showed an obvious role in improving the differentiation, and the early stimulation (day 2-5) in cell culture period of 8 days had more signicant contribution to promote cellular osteogenic differentiation. These are attributed to activating of an appropriate VGIC and intracellular calcium ion oscillation, which induces up-regulation of the expression of osteogenesisrelated gene and subsequently triggers osteogenic differentiation. Therefore, this study provides a way to regulate the osteogenic differentiation through electrical stimulation in the view of charge injection, a guide to design electrode in considering materials with C Q and an alternative insight into understanding of the mechanism of electrical stimulation to cellular responses.

Conflicts of interest
There are no conicts to declare.